Rieke, FredFreedland, Julian2023-08-142023-08-142023-08-142023Freedland_washington_0250E_25681.pdfhttp://hdl.handle.net/1773/50186Thesis (Ph.D.)--University of Washington, 2023The human visual system relies on neural signals that travel along the optic nerve from the eye to the brain. Signals within the optic nerve are generated by individual neurons (“ganglion cells”) that reside within the retina and produce neural spikes in response to specific visual inputs. Establishing a causal link between the visual features in a single scene and the subsequent spike response of retinal ganglion cells is an important step in developing a complete understanding of human vision – and can aid the design of prosthetic eyes that restore sight. In this dissertation, we focus our efforts towards how retinal ganglion cells in the mammalian retina encode realistic scenes from the natural world (“natural scenes”). Natural scenes are typically difficult to study in the retina due to their high dimensionality (large number of pixels) and complex statistical structure (how pixels are arranged). In Chapter 2, we gain traction on this problem by simplifying the spatial structure of natural movies and asking, using electrophysiological techniques, whether these simplifications are reflected in the spike responses of retinal ganglion cells. We find that in parasol ganglion cells, natural movies with 16 wedge-shaped pixels elicit similar responses to the original natural movie. Importantly, these simplified stimuli perform consistently across populations of neurons and only require knowledge of the neuron's classical receptive field, enabling relatively accurate predictions of neural responses to natural scenes with minimal inputs. In Chapter 3, we utilize a similar electrophysiological approach to investigate whether retinal ganglion cells can resolve certain spatial disruptions to natural scenes. We find that breaking the spatial structure of images (introducing “spatial discontinuities”) uniformly delays spike responses. Our work provides insight into how complex visual inputs are encoded by the mammalian visual system.application/pdfen-USCC BYdimensional reductionencodingretinaretinal ganglion cellssensory systemsvisionNeurosciencesBiophysicsMolecular engineeringThesis